Messy Data: Teaching Students Early on About the Realities of Data

• Activity Number: 408
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Education
• Author(s): Ann Cannon*
• Companies: Cornell College
• Keywords: data science ; education ; data cleaning

Abstract

Students in introductory statistics courses typically only see "nice" data sets - data sets that are already organized in a fashion conducive to analysis and with no faulty data. But the reality today is that many data sets do not start out this way. The new ASA undergraduate curriculum guidelines suggest that we should be infusing our undergraduate curriculum with more data science so that students can, in fact, work with messy data. I will be reporting on an interdisciplinary topics course that I co-taught this year with a computer science colleague that focused on data cleaning, data visualization, and touched on some issues with big data.